Statistical Issues in analysing health-related quality of life data
27 February 2007 14.00
The Royal Statistical Society, 12 Errol Street, London EC1Y 8LX
The use of bootstrap methods in analysing quality of life outcome data
Quality of Life (QoL) measures are becoming increasingly used in
clinical trials. QoL outcomes are usually measured on an ordinal scale.
This scaling means that the measures tend to generate data that have
discrete, bounded and skewed distributions. Thus standard methods of
analysis that assume Normality and constant variance may not be
appropriate. Conventional statistical advice would suggest that
non-parametric methods (such as the bootstrap) be used to analyse QoL
data. This talk will show how we can use the bootstrap for hypothesis
testing, estimation of standard errors and confidence intervals for
parameters in studies that have used QoL outcomes.
Application of the Integrated Quality-Survival Product in Clinical Trials
Quality-adjusted life years is often a relevant outcome measure on which
to compare treatments in clinical trials. The original Q-TWiST model
(Goldhirsh et al 1989) has been used for quality-adjusted survival
analysis but was not devised for the analysis of longitudinal quality of
life data, now commonly collected in clinical trials. This talk will
describe a more appropriate methodology for this scenario called the
integrated quality-survival product, illustrating the approach using
data from a pancreatic cancer clinical trial.
No Registration is required for this event
Meeting Contact: Gordon Taylor ( [log in to unmask])
Organising Group(s): Royal Statistical Society Medical Section
Dr Stephen Walters
Senior Lecturer in Medical Statistics
Medical Statistics Group
School Health and Related Research (ScHARR)
University of Sheffield, Regent Court,
30 Regent Street, Sheffield S1 4DA.
T: +44 (0)114 2220730